Advised by professor Chris Atkeson and Akihiko Yamaguchi, our group of 6 set out on a semester long project to fabricate a soft robotic actuator. Our intention was to build a prototype finger assist that could be used to help elderly people with limited grip strength hold and move objects in their daily life, and explore and learn about soft robotics.
As the only mechanical engineer on a team with two electrical engineers, two computer scientists, and a human-computer interactionist, a lot of my responsibilities were in fabrication.
There are four basic finger grip types and we chose to model after the full fist, as it required the most amount of "joints," and accordingly would be an interesting challenge to pursue. From there, we began fabrication of our mold. It took several iterations to reach a two part 3D printed plastic mold. Our biggest challenge was creating the dimensions we needed in the internal cavities, as spotting bubbles in ecoflex can be fairly challenging.
Several attempts at molding the epoxy later, we learned what techniques work, and what don't, and yielded three suitable actuators! After sealing together the two halves of the soft actuators and testing their airtightness, it was time to build the array of valves that would actuate the actuator. Using a variety of tubing, air pumps, solenoid valves, and an Arduino, I began building an array of valves so that the soft actuator could be inflated to a specified pressure at the push of a button, and released at the button's depress.
After our computer engineer group mates finished the software part of our system, it was time for us to integrate it all together. As with any good project, almost nothing went right when we first tried getting it all working. We ended up needing to find a new tube of a slightly smaller diameter, and needed a faster drying silicone spread to seal parts of the solenoid valves.
Crunch time, two weeks before due date
Initial silicone casts
Four basic finger grip types (full fist, hook fist, straight fist, and platform).
Custom 3D printed silicone molds
A radio telescope is an astronomy device used to observe the radio emissions of various celestial bodies; this is known as radio astronomy. A radio telescope can give us interesting and different information from what we see in the visible spectrum (as an optical telescope does) by observing what goes on in the radio spectrum.
This summer I had the goal of building a simple radio telescope to observe the Hydrogen Line. The Hydrogen line is the name of the specific spectral line created by neutral hydrogen atoms in space as they change energy levels. The electromagnetic radiation emitted to create the spectral line occurs at 1420.405MHz, this region of microwave is resilient enough to the conditions on the way to earth such that they are the most commonly observed spectra in amateur Radio Astronomy.
The hardware required to complete this project included: • Raspberry Pi 3 • USB Software Defined Radio • Low Noise Amplifier • 2 Line Amplifiers • A Coaxial Power Source • Several Coaxial Cables/Adapter • Coaxial Power Inserter • 18in Satellite Dish • Power Strip/Adapters
In order to actually gather and transform the unprocessed radio data, I primarily used the Linux based command line tool Radio Astronomy FFTW, and secondarily used other open source command line tools from the GNU-Radio project.
One piece of data gathered from my telescope are spectrograms like the one below of the hydrogen line from several consecutive 30 minute scans between 1420.381MHz and 1420.431MHz. Click here to load an animated version. (once it loads, click a few times in the black window to begin the animation)
The most awesome and interesting data plot: signal power around the frequency of the hydrogen line for a given area of the sky as the earth rotated and moved in orbit over a 15 hour period.
All the required hardware
spectrogram of the hydrogen line from several consecutive 30 minute scans between 1420.381MHz and 1420.431MHz.
A Bravit Candle is a style of candle with a wick that twists to move the flame in artistic patterns or shapes, getting more beautiful as the candle is burned. They are usually designed for use in important romantic dinners, like a proposal. I thought it would be fun to make a version of the candle that looks nice to start, but then ends up a sad scary mess, perhaps more for a dinner that ends with a break up rather than proposal.
The candle began with paraffin wax, wick, a little bit of vanilla scent, and a mold I made out of wax paper and cardboard. I set the wick in the mold each at a different length and at some points twisted around each other to ensure a seemingly unpredictable and interesting flame pattern. Then I brought the wicks up through the top of the candle looking normal and only the slightest bit interesting.
And then i lit the candle, and it got a whole lot more exciting…
When it cooled and I removed it from the mold, this is what came out!
Then it was time to light the candle and see what happens, definitely looking normal to start.
After about 30min, it was looking a little more chaotic, so i put it on some paper to keep the mess off the table.
A few hours in, it was looking a little more exciting.
And finally, a big awesome mess!
Really, just, a huge mess.
48 Hour Chair
In Design for Manufacturing and The Environment class, we were given the task of building a chair capable of supporting 250lbs. We had 1 week between the assignment out date and due date, which meant, due to package delivery problems, that I had 48hours to build a chair and create a report with engineering drawings and design rational.
My inspiration for the chair was the Zig Zag Chair, an outline of which is above. Versions of this chair sell for many hundreds if not thousands of dollars, and I wanted to see if I could do it using as much scrap materials, recycled parts, and sustainable manufacturing principals as possible.
I decided I wanted to make an extremely robust chair, capable of supporting a large amount of weight, while targeting three specific DFM principals the most: reducing part count, ease of disassembly, and ease of recyclability. To meet my goals in those three areas, I decided I would go with an iron stock body, and let the rest of the design be informed by the material I was able to gather.
I began thinking of materials I could find for the seat part of the chair that would look good and support weight. I settled on recycled wood flooring, which a Home Depot employee kindly gave me an old sample of. As I now had less than 36 hours left, I began machining the metal for the frame of the chair.
In order to reduce part count, I opted for no mechanical fastening, and instead chose to weld the body of the chair together. While I was just using an old MIG machine the DFM intention would be that with only 45 degree and 90 degree angles, and clamping fixture could easily be build to allow a TIG welding robot to easily build the chair in a high volume environment. While welding can be a fairly dirty manufacturing process, the durability gain, the lack of any mechanical fastener or adhesive, and the fact that TIG welding can be done fairly cleanly, made this a worthwhile trade off in my my mind. Having not done any welding for about a year, I was mostly happy with how my joints turned out. Just ignore the monstrosity that is the joint on the upper left...
After machining the right angles in the machine shop, bringing the stock to the welding studio on the other side of campus, and then bringing the body back to the machine shop, I was ready to add the seating.
I looked through machine shop scrap to find something I could use to affix the flooring to the base of the chair, and found some hemp twine. After boring holes into the wood, I threaded the twine and tightened the part into place, doing a similar thing for the backrest of the chair.
In the end by bill of materials came down to 11.6ft of twine, 2.67 square feet of recycled flooring, and 144in of plains steel square stock. At the end of the chair's life, the seat and backrest are easily removed by cutting away the twine. The metal, twine, and wood, can then all be recycled separately.
Zig Zag Chair
My chair, sitting among the other chairs for the class.